import torch
from transformers import AutoTokenizer, AutoModelForCausalLM

# Load model and tokenizer
model_name = "Syko/SykoNaught-v1"
tokenizer = AutoTokenizer.from_pretrained(model_name)
model = AutoModelForCausalLM.from_pretrained(model_name)

def handle(inputs):
    """
    Handle incoming inference requests.
    """
    input_text = inputs.get("inputs", "")
    max_new_tokens = inputs.get("parameters", {}).get("max_new_tokens", 50)
    temperature = inputs.get("parameters", {}).get("temperature", 0.7)

    # Tokenize input
    input_ids = tokenizer(input_text, return_tensors="pt").input_ids

    # Generate output
    output = model.generate(
        input_ids,
        max_new_tokens=max_new_tokens,
        temperature=temperature,
    )
    output_text = tokenizer.decode(output[0], skip_special_tokens=True)

    return {"generated_text": output_text}